Published in bioRxiv, 2024
By using shallow recurrent decoder networks to expand datasets of human motion, wearable sensors could be used for more continuous and less intrusive monitoring of human motion in natural environments.
Download here
Published in arXiv, 2023
The time-history of mobile sensors can be used to encode global information of the measured high-dimensional state space.
Download here
Published in Journal of Biomechanics, 2023
Discrepancy modeling is a unique and innovative tool that complements current biomechanical modeling approaches and may accelerate the discovery of individual-specific mechanisms driving responses to exoskeletons, other assistive devices, and clinical interventions.
Download here
Published in arXiv, 2022
The characterization of the underlying instability has significant implications for engineering design. Specifically, instead of engineering at the level of the complex physical processes, one should instead consider the potential for engineering the overall gain and loss curves.
Download here
Published in SIAM Journal on Applied Dynamical Systems, 2022
As data-driven modeling continues to gain momentum, it is imperative that researchers utilize domain knowledge (e.g., first principles physics) to model complex systems. However, in all disciplines, model- measurement mismatch exists. The lack of investigation into this error leaves the missed opportunity to resolve model-measurement mismatch, disambiguate deterministic effects, and improve the underlying model.
Download here